Problem drinking MSM are at risk for relapse because their social milieux are often associated with alcohol, and are at elevated risk for HIV because of high base rates of HIV and because drinking is associated with HIV-risk behavior. Although substantial research exists on HIV risk reduction in MSM alone or in combination with alcohol abuse treatment, few studies have examined the impact of alcohol-only treatment on HIV risk. Problem drinkers are most receptive to interventions focused on moderation of drinking rather than abstinence; however moderation-oriented cognitive-behavior therapy (CBT) is limited in that success depends on the successful implementation of new coping strategies for urges to drink and development of self-efficacy for coping. Naltrexone (NTX) is also a well established treatment for problem drinkers with moderated drinking goals, but its effects diminish after treatment is discontinued. The limitations of these treatments suggest that their combination might enhance the efficacy of both. We propose combining moderation-oriented CBT with NTX in the treatment of problem drinking,HIV- positive and -negative MSM, to evaluate their efficacy alone and in combination. We also propose to utilize new data collection technology, Interactive Voice Response, to collect data on daily relations among drinking, sexual behavior and psychological variables thought to mediate treatment response. Our objectives are to: 1) enroll a sample of 240 problem drinking MSM using active and passive recruitment strategies; 2) evaluate the efficacy of 12 weeks of randomly assigned treatment, with 50 mg of NTX or placebo, combined with modified, behavioral self-control therapy or brief supportive therapy; 3) evaluate conditional relationships between heavy drinking and likelihood of HIV risk behavior; 4) evaluate daily associations among mood, craving, self-efficacy, motivation, and drinking. Assessments will include baseline, 3, 6, & 9 month follow-up. This application is of tremendous relevance to alcoholism treatment and HIV prevention. First, we will evaluate established treatments for problem drinking in an under-studied but unique population, namely MSM. Second, we will examine the impact of alcohol-focused treatment on HIV risk behavior in a very high- risk population. Third, we will use new data collection technology to examine complex relationships among drinking, HIV risk, and psychological factors to better inform future interventions.